Search

Cataracts are a leading cause of blindness worldwide. Surgical removal of cataracts is a safe and effective procedure to restore vision. However, a large number of patients later develop vision loss due to regrowth of lens cells and subsequent degradation of the visual axis leading to visual disability. This postsurgical complication, known as posterior capsular opacification (PCO), occurs in up to 30% of cataract patients and has no clinically proven pharmacological means of prevention. Despite the availability of many compounds capable of preventing early steps in PCO development, there is currently no effective means to deliver such therapies into the eye for a suitable duration. To model a solution to this unmet medical need, we fabricated acrylic substrates as intraocular lens (IOL) mimics scaled to place into the capsular bag of the mouse lens following a mock-cataract surgery. Substrates were coated with a hydrophilic crosslinked acrylate nanogel designed to elute Sorbinil, an aldose reductase inhibitor previously shown to suppress PCO. Insertion of the Sorbinil-eluting device into the lens capsule at the time of cataract surgery resulted in substantial prevention of cellular changes associated with PCO development. This model demonstrates that a cataract inhibitor can be delivered into the postsurgical lens capsule at therapeutic levels.

Kinetic and structural changes in recombinant human aldose reductase (AR) due to modification by S-nitrosoglutathione (GSNO) were investigated. Incubation of the enzyme with 10-50 microM GSNO led to a time- and concentration-dependent inactivation of the enzyme, with a second-order rate constant of 0.087 +/- 0.009 M-1 min-1. However, upon exhaustive modification, 30-40% of the enzyme activity was retained. The non-inactivated enzyme displayed a 2-3-fold change in Km for NADPH and Km fordl-glyceraldehyde, whereas the Km for the lipid peroxidation product, 4-hydroxy-2-trans nonenal (HNE), was comparable to that of the untreated enzyme. The residual activity of the enzyme after GSNO treatment was less sensitive to inhibition by the active site inhibitor sorbinil or to activation by sulfate. Significantly higher catalytic activity was retained when the enzyme was modified in the presence of NADPH, suggesting relatively low reactivity of the E-NADPH complex with GSNO. The modification site was identified using site-directed mutants in which each of the solvent-exposed cysteines of the enzyme was replaced individually by serine. The mutant C298S was insensitive to GSNO, whereas the sensitivity of the mutants C303S and C80S was comparable to that of the wild-type enzyme. Electrospray ionization mass spectroscopy of the GSNO-modified enzyme revealed a major modified species (70% of the protein) with a molecular mass that was 306 Da higher than that of the untreated enzyme, which is consistent with the addition of a single glutathione molecule to the enzyme. The remaining 30% of the protein displayed a molecular mass that was not significantly different from that of the native enzyme. No nitrosated forms of the enzyme were observed. These results suggest that inactivation of AR by GSNO is due to the selective formation of a single mixed disulfide between glutathione and Cys-298 located at the NADP(H)-binding site of the enzyme.

Aldose reductase is an NADPH-dependent oxidoreductase that catalyzes the reduction of a variety of aldehydes and carbonyls, including monosaccharides. Intense interest in the discovery and characterization of inhibitors has developed since the action of this enzyme has been linked to the pathogenesis of some diabetic complications. Since past studies indicated that most inhibitors act noncompetitively or uncompetitively versus substrate in the direction of aldehyde reduction, it was assumed that they bind at one or more sites distinct from the active site. However, the crystal structure of aldose reductase complexed with inhibitor [Wilson et al. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 9847-9851] unambiguously revealed the inhibitor bound at the active site. The present study was undertaken to address this apparent discrepancy. Using a fluorometric assay, it was determined that zopolrestat, an acetic acid-type inhibitor, bound to aldose reductase complexed with either NADPH or NADP+. In contrast, the spirohydantoin-type inhibitor sorbinil demonstrated preferential binding to the binary enzyme.NADPH complex. Prior incubation of the enzyme.NADPH complex with zopolrestat prevented subsequent sorbinil binding. These results, together with the published structure of the ternary enzyme.NADPH.zopolrestat complex, are consistent with the conclusion that both sorbinil and zopolrestat bind at the active site. We propose that mixed inhibition patterns previously observed with sorbinil are due to inhibitor binding to both structural isomers of the enzyme.NADPH complex. Similar patterns for inhibition by zopolrestat are due to tight binding of the inhibitor. Substrate inhibition in the direction of aldehyde reduction occurs as a result of substrate binding to the enzyme.NADP+ complex.

This study was undertaken to compare the ability of two guanidine compounds (aminoguanidine and methylguanidine), with different in vitro effects on NO synthase activity and AGE formation, to inhibit diabetic vascular dysfunction developing early after the onset of diabetes. In rats with STZ-induced diabetes of 5-wk duration, regional vascular [125I]albumin permeation was increased about two- to threefold in ocular tissues, sciatic nerve, and aorta; in general, both guanidine compounds normalized albumin permeation in diabetic rats without affecting it in controls. Methylguanidine was only approximately 7% as effective as aminoguanidine as an inhibitor of AGE formation from L-lysine and G6P; both compounds were poor inhibitors of AR. Methylguanidine was approximately 1-5% as potent as aminoguanidine and L-NMMA as an inhibitor of the cytokine- and endotoxin-inducible isoform of NO synthase. In contrast, the potency of methylguanidine as an inhibitor of the constitutive isoform of NO synthase was comparable to that of aminoguanidine, and both guanidine compounds were much less effective than L-NMMA. These observations suggest a role for a relative or absolute increase in NO production in the pathogenesis of early diabetic vascular dysfunction and raise the possibility that inhibition of diabetic vascular functional changes by aminoguanidine may reflect inhibition of NO synthase activity rather than, or in addition to, prevention of AGE formation.

We have resolved and characterized isoforms of aldose reductase from bovine and porcine lenses by preparative isoelectric focusing with narrow pH gradients using the Rotofor. Both bovine and porcine lens aldose reductases were resolved as two enzyme isoforms. The bovine isoforms were Mr40400 +/- 445 polypeptides of pI4.71 and 5.19. Porcine isoforms were Mr41500 +/- 450 polypeptides of pI 4.90 and 5.30. Staphylococcus aureus V-8 protease digestion patterns for each set of isoforms were essentially identical and all isoforms probably contain blocked amino terminal amino acids. Antiserum to bovine lens aldose reductase cross-reacted with porcine lens aldose reductase. Each isoform displayed substrate preferences characteristic of mammalian aldose reductases. With purification, both bovine and porcine lens aldose reductases became less sensitive to inhibition by 6-fluoro-spiro-(chroman-4.4'-imidazolidine)-2',5'-dione (sorbinil).

The effect of aldose reductase inhibitors such as sorbinil, alrestatin, and quercitrin has been studied on the aldose reductase purified from human brain and lens, and aldehyde reductase I purified from human liver, and aldehyde reductase II purified from human brain, liver, and red cells. None of the aldose reductase inhibitors have been found to be specific for aldose reductase. Fifty micromolar sorbinil besides inhibiting aldose reductase, completely inhibits aldehyde reductase II from the brain, liver and red cells. Similarly, alrestatin and quercitrin also are potent inhibitors of aldehyde reductase I and aldehyde reductase II.